The incorporation of increasing numbers of discrete devices into progressively smaller integrated circuits remains an important challenge in the manufacture of Very Large Scale Integration (VLSI) structures. For example, the implementation of CMOS technology into integrated circuits usually involves imparting a prescribed set of manufacturing attributes to the integrated circuit, which may include selected dopant concentrations, channel lengths, interconnect dimensions, contact shapes, or other pertinent attributes, which collectively permit the integrated circuit to provide a desired function.
Many of the desired features in VLSI structures may be formed using photolithographic methods. Briefly, and in general terms, a photolithographic mask (or reticle) is formed that includes a desired pattern corresponding to a particular masking step for the structure. The pattern generally includes optically transparent areas and optically opaque areas that are suitably arranged on an optically transparent supporting substrate. The mask may then be interposed between an illumination system and a layer of an illumination-sensitive photoresist material applied to a semiconductor wafer. The illumination system projects illumination radiation through the optically transparent portions of the mask and onto the photoresist material, which suitably changes the properties of the photoresist material. Subsequent development of the exposed photoresist material thus permits the selective differentiation between exposed and non-exposed areas in the photoresist material so that the desired pattern may be subsequently formed on the semiconductor wafer.
When a wavelength of the illumination radiation is greater than a minimum feature size expressed on the mask, various optical effects may adversely affect the quality of features formed on a semiconductor structure. For example, edges between transparent areas and opaque areas generally leads to diffractive effects, which generally causes constructive interference when the waves of the illumination radiation are bent and re-radiated, resulting in exposure reduction in areas corresponding to the transparent areas of the mask, and undesired illumination in areas corresponding to opaque portions of the mask. As feature densities in semiconductor structures increase (and correspondingly, feature sizes decrease), diffractive effects, as well as other optical effects become more prominent limiting factors in photolithography.
Accordingly, various compensation methods are available that may increase the pattern fidelity in the structure. For example, in one known method, optical proximity correction (OPC) may be used to perturb the shapes of transmitting apertures, or other features on the mask to enhance optical resolution in the sub-wavelength regime. In general, the perturbed features on the mask are sub-resolution features since they are generally not printed onto the structure during the exposure process. Accordingly, these features are collectively referred to as sub-resolution assist features. Examples of sub-resolution assist features may include “serifs” to reduce corner rounding in the features formed in the structure, and “hammerheads” to reduce the shortening of end line features. Other sub-resolution assist features may include scattering bars, or “outriggers”, and “inriggers” that improve line width control in the structure. Still other methods may be used to improve the resolution of features in the sub-resolution regime. For example, Phase Shift Masking (PSM) methods generally enable transparent regions on the mask to transmit phase-shifted illumination to the structure in order to reduce destructive interference that may occur between transparent areas that are separated by an opaque area on the mask. Still other methods may be directed to the illumination system itself. For example, an incident radiation angle (σ) and/or the numerical aperture (NA) of a projection lens may be suitably configured to resolve relatively dense lines and spaces.
Although the foregoing methods constitute improvements in the state of the art that permit aggressive reductions in feature size, still other arrangements of features may introduce resolution problems that are not fully addressed by the foregoing methods. For example, desired features that include isolated and semi-isolated regions, such as lines or spaces, may not be adequately addressed by the foregoing methods. Therefore, what is needed in the art are sub-resolution assist methods and structures that permit the formation of these features.